top of page

Practical tips for data visualization and storytelling.

Join over 800 people who receive tips every month

Thanks for submitting!


2020 Favorites: The Year’s Best Data Visualizations

To say that 2020 was an eventful year would be an understatement. The global pandemic, the economic roller coaster, the Black Lives Matter movement, and the US presidential elections were not just events that will forever be mentioned in history books. These were events that brought data and charts to the forefront. Regardless of your age, profession, or location, you were likely exposed to charts on a daily basis in 2020.

Trend line by year - 2021 stands out being the highest

To cap off this long and unusual year, we selected our top 10 favorite visualizations (in no particular order!). With so many data visualizations to choose from, our job was not easy. We narrowed down the list based on 3 criteria:

TIMELINESS = Were the data and the point in time when the graph was published relevant?

UNIQUENESS = Did the chart bring a new perspective? Or, was the chart so innovative that it grabbed our attention right away?

BEST PRACTICES = Did the graph follow the data visualization principles?


NYT front page with 100,000 names

This is not your typical data visualization. In fact, one could argue that this is not a data visualization at all. Well...we'd like to argue otherwise. Datum is defined as "a single piece of information." Words and letters are information, just as much as numbers are information. And 100,000 names of people who died due to COVID-19 certainly make for an unforgettable data visualization.

Is this New York Times cover page timely? Absolutely. It was published shortly after the death toll reached 100,000 deaths in the U.S. Is it unique? There's no doubt that it grabbed my attention right away. Does this visualization follow data visualization best practices? One could argue that this NYT front page looks cluttered and that there's no key data point that stands out. We believe that this visualization is a perfect example of why rules should only be the starting point.


Learn more about our virtual data storytelling and visualization training for your team!


Douïeb's map came in response to the most common way of representing the US presidential elections - a filled map. The problem? Such maps highlight the geographical areas instead of the electoral importance. The option proposed here - representing each county by a dot sized proportionally to the amount of its associated voters - paints a much more accurate picture.

This map was timely published, as the votes were still being counted and dozens(if not hundreds) of maps were published on a daily basis. Was this visualization unique? It certainly stood out and challenged the way the US presidential elections were visually represented. Did this map follow best practices? It did! It displayed the data more accurately than a filled map.

Cook County, IL - map with a radius of 1.3 miles

This visualization is truly unforgettable, partially because it is so personalized. In fact, we are such big fans of this interactive graph, that we even covered it in a recent YouTube video.

Is this visualization timely? Absolutely. It was published as the COVID-19 death toll in the US was rapidly increasing. Is it unique? We believe it is. It creates a hypothetical scenario of how the death toll would have looked like, had it happened in close proximity to each one of us. Does this visualization incorporate best practices? It does - one design element that stood out the most was how the monochromatic scheme helped paint a heartbreaking picture.

#4: The New York Times: India Pollution Inequality

Two kids - Monu and Aamya and the levels of pollution (148.9 and 36.6, respectively)

Unlike the other graphs that we selected, this one doesn't cover COVID-19 or the US presidential elections. Nevertheless, the topic and the data are equally important - air pollution in New Delhi, India. For this article and graphs, the New York Times staff worked with pollution researchers to collect the data. The story follows a day in the life of Monu and Aamya, two Indian kids who live in the same city but are regularly exposed to different levels of pollution, due to their socioeconomic class.

Is this visualization timely? We believe it is. While the world has focused on a different health topic this year (the COVID-19 pandemic), the issue of high levels of pollution in India is critical. Are these New York Times charts unique? They are - as you scroll through the article, the charts are interactive and overlay the images. Does this visualization follow best practices? It does. What stood out the most for us was the use of annotations, which provided helpful additional context.

#5. McKinsey & Co.: The Emotion Archive

Bubble charts with the emotion index by country

This is one of our favorite COVID-19 data visualizations. In fact, we liked it so much that we made a video about it. This McKinsey & Co. visualization is asking the question: “How have people around the world coped with the COVID-19 crisis?” The visualization starts at a global level and, as you scroll down, you see the most prominent emotions by country.

Are these charts timely? Absolutely. They explore the pandemic at a global level. Are they unique? We believe they are. While most COVID-19 data visualizations have featured the trends in number cases and deaths, these McKinsey & Co. charts tell a different story - that of the emotional impact. Does this visualization follow best practices? It does. Among many principles that we could mention, one that is immediately apparent is the use of personal stories to truly bring the data to life.

Area charts with one chart per state, shwoing the votes after polls closed. Zoom into GA and NC.

FiveThirtyEight published a ton of amazing charts in 2020. This specific one depicts the timeline of vote tallies for each candidate after polls closed, by state. The visualizations starts off with a legend, which clearly explains how the readers should interpret the area charts. Each area chart represents a US state and follows each state's specific geographic location.

Is this graph timely? Given that it was published at a time when everyone was anxiously awaiting for all states to finalize the vote count, this visualization was very well timed. Is it unique? It is, as most US election charts displayed the data as a filled map. An area map was a somewhat unexpected (but effective) choice. Finally, what data visualization best practices stood out? Zooming in and annotating the corners of the NC and GA area charts. Without zooming in and annotating, these two critical states would have gone unnoticed.

Bee swmarm showimg that Biden is likely yo win the election  with 89 out 100 vs 10 out of 100 for Trump.

We first noticed this chart on Twitter, back in July, when @wiederkehra (Sr. Visual Journalist at FiveThirtyEight) published a behind-the-scenes on Twitter. The visualization shows that Biden was favorite to win the US elections, based on a model that included 40,000 simulations and a sample of 100 outcomes.

Tweet showing Biden and Trump. Who's winning the election?

Is this visualization timely? Undoubtedly! FiveThirtyEight published this graph as the US was preparing for the US presidential elections. Was it unique? We still remember it, although it was published over 5 months ago. Two elements made this graph memorable: the use of art (Trump and Biden illustrations), and the chart choice (bee swarm plot). Is this graph following data visualization best practices? While there are many principles that we could mention here, we'll focus on two in particular - the excellent use of color (not your typical shade) and the bee swarm plot, which is a great chart choice to highlight the likelihood for Biden to win.

Line of people waiting to get the vaccine.

This chart was published towards the end of 2020. It starts off with the question: "A vaccine may be around the corner, but how long will it be until you get the shot?" You are asked to input your age, location, and COVID-19 health risk. Then, hit submit, and ... a visualization comprised of 100 people shows you what your place in line is for getting a vaccine.

Is this visualization timely? Definitely. Now that vaccines are being produced and distributed, "How long will it be until I get the shot?" is a question that's top of mind. Is this visualization unique? It absolutely stands out and helps each of thoroughly envision our place in line. Does it follow visualization best practices? We believe that the minimalist approach is effective at keeping the readers' focus on each person in line as they scroll down the page.

Two bar charts accompanied by a line trend - new cases per day and new deaths per day from 3/1 to 12/26. Both spiked in the last weeks.

We know...this looks like a common chart. It doesn't stand out as much as the other visualizations that we cover in this blog post. Yet, choosing a more common chart type can actually be a good and effective decision. In this case, USA Today shows trends over time for COVID-19 new cases and new deaths per day.

Are these graphs timely? Undoubtedly. They cover a topic that's been in the news since early 2020. Are they unique? While the chart type is not unique, the more nuanced design choices stand out - colors and mix of bars and line graphs to display trends. Are the graphs following key best practices? They are - design elements such as spare labeling on the X and Y axes and color choices create powerful and effective data storytelling.

Mona Chalabi is known in the world of data visualization for her unique style. In this carousel-type visualization, she depicts the difference between the different outcomes when police officers killed people in the US...with an overwhelmingly large majority resulting in no charges whatsoever.

Is this visualization timely? Definitely - it was published in the midst of the Black Lives Matter movement. Is it unique? The artistic aspect, as well as the carousel structure create a memorable and outstanding visualization. Is the graph following best practices? The gradient color scheme and label proximity are two data visualization best practices that stand out immediately.


As you can imagine, we are not the only ones to pick our favorite charts of the year. With so many charts to chose from, other practitioners have also selected their favorite data visualizations. Here are a few:

Celestial visualization in two circles that partally overlap.

Line graph showing cumulative deaths in Engald due to COVID-19 from March to October.

Global temperature change (1850 -2019) - barcode plot going from blue to red.


bottom of page